Neuromorphic visual perception, by emulating the efficient information processing mechanisms of biological vision systems and integrating innovations in materials and device architectures, offers novel solutions for artificial intelligence sensing. For instance, the incorporation of low-dimensional materials (e.g., quantum dots, carbon nanotubes, and two-dimensional materials) optimizes device optoelectronic properties, while the synergistic design of organic semiconductors and oxide materials balances flexibility with complementary metal-oxide-semiconductor (CMOS) compatibility. Representative neuromorphic devices such as memristors and neuromorphic transistors address traditional vision system bottlenecks via near-sensor and in-sensor architectures in data transmission latency and energy consumption, offering a new paradigm for highly integrated, energy-efficient real-time perception. However, critical challenges—including device non-uniformity caused by material interface defects, system instability induced by memristor conductance drift, and environmental adaptability under complex illumination—remain barriers to scalable applications. This review comprehensively examines neuromorphic visual perception devices from the perspectives of device structure, operational mechanisms, materials, and applications. It explores the pivotal roles of memristors, electrolyte-gated transistors, and other neuromorphic devices in optical signal perception and information processing, with a focus on their implementations in visual perception tasks and future prospects.

Neuromorphic computing is a brain-inspired computing paradigm that aims to construct efficient, low-power, and adaptive computing systems by emulating the information processing mechanisms of biological neural systems. At the core of neuromorphic computing are neuromorphic devices that mimic the functions and dynamics of neurons and synapses, enabling the hardware implementation of artificial neural networks. Various types of neuromorphic devices have been proposed based on different physical mechanisms such as resistive switching devices and electric-double-layer transistors. These devices have demonstrated a range of neuromorphic functions such as multistate storage, spike-timing-dependent plasticity, dynamic filtering, etc. To achieve high performance neuromorphic computing systems, it is essential to fabricate neuromorphic devices compatible with the complementary metal oxide semiconductor (CMOS) manufacturing process. This improves the device’s reliability and stability and is favorable for achieving neuromorphic chips with higher integration density and low power consumption. This review summarizes CMOS-compatible neuromorphic devices and discusses their emulation of synaptic and neuronal functions as well as their applications in neuromorphic perception and computing. We highlight challenges and opportunities for further development of CMOS-compatible neuromorphic devices and systems.